DSRC与车载传感器融合的智能车辆目标跟踪方法研究

发布时间：2017-12-27 21:39

本文关键词：DSRC与车载传感器融合的智能车辆目标跟踪方法研究　出处：《重庆邮电大学》2016年硕士论文　论文类型：学位论文

更多相关文章： 智能车辆 目标跟踪 DSRC 数据关联

【摘要】：智能车辆目标跟踪系统用于对周围目标进行实时跟踪,准确地获取目标状态信息是安全预警、辅助驾驶和自动驾驶等系统必要的前提条件。在协同式智能车辆场景中,传统的车辆跟踪方法单纯地依靠车载传感器对目标进行跟踪,没有充分利用通过车间协同方式得到的信息,存在数据关联和跟踪滤波结果准确性不高的问题,因此,研究利用通过车间协同方式获取信息的目标跟踪方法具有重要的意义。本文针对现有的车辆目标跟踪方法在协同式智能车辆场景中存在的不足,提出一种利用专用短程通信技术(Dedicated Short Range Communications,DSRC)与车载传感器融合的智能车辆目标跟踪方法。在协同式智能车辆场景下,本文方法考虑了协同目标车辆可通过DSRC通信技术发布其自身的状态及身份等信息,利用这部分信息对数据关联和跟踪滤波的结果进行修正,有效的提高了数据关联准确性和跟踪精度。本文方法的基本思路是:首先主车通过激光雷达传感器采集数据并对数据进行初步处理,利用数据关联算法对激光雷达的观测结果进行数据关联;其次,通过DSRC通信设备接收并处理协同目标车辆发布的自身状态及身份等信息,利用处理后的DSRC信息对数据关联的结果进行修正;最后对修正后的数据关联结果进行跟踪滤波并与目标车辆发布的状态信息进行融合。通过仿真对比,验证了本文方法的有效性。本文将提出的跟踪方法应用到协同式智能车辆场景的目标跟踪中,设计并开发了智能车辆目标跟踪系统。该系统主要包含数据采集处理和目标跟踪两大模块,数据采集处理模块负责完成对各车载传感器的数据采集与处理,将处理后的数据传给车辆目标跟踪模块进行目标跟踪,最后将跟踪等结果通过人机交互界面实时显示。本文设计实现的智能车辆目标跟踪系统按照高内聚,低耦合的思想对软件进行模块化处理,使得该系统具有良好的可扩展性。最后,对本文开发的智能车辆目标跟踪系统进行了实车实验,通过实验结果验证了本文方法和系统的有效性。
[Abstract]:The intelligent vehicle target tracking system is used for real-time tracking of the surrounding targets, and accurately obtain the information of the target status. It is a necessary prerequisite for such systems as safety early warning, auxiliary driving and automatic driving. In Cooperative Intelligent Vehicle in the scene, the traditional vehicle tracking method simply rely on on-board sensors to track the target, did not make full use of the information obtained through the way of collaborative workshop, existing data association and tracking filtering accuracy is not high, therefore, has an important significance to research the use of tracking method to obtain information through the way of collaborative workshop the target. Aiming at the shortcomings of existing vehicle target tracking methods in collaborative intelligent vehicle scene, this paper proposes an intelligent vehicle tracking method based on fusion of Dedicated Short Range Communications (DSRC) and vehicle sensors. In Cooperative Intelligent Vehicle scene, this method considers the cooperative target vehicle can be released and its own identity and other information through the DSRC communication technology, the use of this part of the information on the data association and tracking filter was modified, and effectively improve the accuracy of the data association and tracking accuracy. The basic idea of this method is: firstly, the main vehicle through the data acquisition of laser radar sensor and preliminary data processing, observation of laser radar data correlation based data association algorithm; secondly, through its own status and identity information such as DSRC communication equipment for receiving and processing the vehicle collaborative target release, the use of DSRC after treatment the results of data association information is modified; at the end of the data association after the correction of state information tracking and released with the target vehicle integration. The effectiveness of this method is verified by simulation comparison. In this paper, the proposed tracking method is applied to the target tracking of collaborative intelligent vehicle scene, and the intelligent vehicle target tracking system is designed and developed. The system includes data acquisition processing and target tracking of two modules, data acquisition module is responsible for data acquisition and processing of the vehicle sensor, the processed data is transmitted to the vehicle target tracking module of target tracking, the tracking results through the man-machine interface Real-Time display. The intelligent vehicle tracking system designed and implemented in this paper modularized the software according to the idea of high cohesion and low coupling, enabling the system to have good expansibility. Finally, the actual vehicle experiment is carried out on the intelligent vehicle target tracking system developed in this paper, and the effectiveness of the method and system is verified by the experimental results.
【学位授予单位】：重庆邮电大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U463.6;TP212

【参考文献】